Convolutional Neural Networks

Convolution Operation

As padding size increases, convolution gets more information, especially at the edges.

The attributes of padding:

Stride refers to the number of steps the filter moves each time.

The attributes of stride:

Pooling is a widely used technique that mainly focuses on reducing the resolution of the output.

Define: * image size: $(H, W)$ * filter size: $(F H, F W)$ * padding: $P$ * stride: $S$

Then the output size: $(O H, O W)$

\begin{array}{r} O H = \frac{H + 2 P - F H}{S} + 1 \\ O W = \frac{W + 2 P - F W}{S} + 1 \end{array}